Mental healthcare utilisation among Danish formerly deployed military personnel and their civilian counterparts: a cohort study

ABSTRACT

Background: Prior studies comparing the mental healthcare utilisation (MHU) of Danish formerly deployed military personnel (FDP) with the general population have not included data on psychotherapy through the Defence or talking therapy with the general practitioner. This study included these and several other data sources in a comprehensive comparison of MHU between Danish FDP and civilians.

Methods: First-time deployed military personnel (N = 10,971) who had returned from a mission to Kosovo, Afghanistan, Iraq or Lebanon between January 2005 and July 2017 were included. A sex and birth-year-matched civilian reference group was randomly drawn from the entire Danish non-deployed population (N = 253,714). Furthermore, a sub-cohort, including male FDP and civilians deemed eligible for military service, was defined. These cohorts were followed up in military medical records and registers covering the primary and secondary civilian health sectors from 2005 to 2018, and the rates of MHU were compared.

Results: Approximately half of the initial help-seeking for FDP took place through the Defence (49.4%), and the remainder through the civilian healthcare system. When help-seeking through the Defence was not included, MHU was significantly lower among FDP in the main cohort during the first two years (IRR = 0.84, 95% CI: [0.77, 0.92]) compared to civilians. When help-seeking through the Defence was included, MHU was significantly higher among FDP compared to civilians both in the first two years of follow-up (IRR = 2.01, 95% CI: [1.89, 2.13]) and thereafter (IRR = 1.18, 95% CI: [1.13, 1.23]). In the sub-cohort, these differences were even more pronounced both in the first two years of follow-up and thereafter.

Conclusions: MHU was higher among Danish FDP compared to civilians only when data from the Defence was included. The inclusion of data on both civilian and military healthcare services is necessary to evaluate the full impact of deployment on MHU among Danish FDP.

HIGHLIGHTS

• This study compared mental healthcare utilisation among Danish deployed military personnel and civilians.

• Most personnel sought help first through the Defence.

• When all data sources were included, mental healthcare utilisation was significantly higher among military personnel.

Antecedentes: Estudios anteriores que compararon la utilización de la atención de salud mental (UASM) del personal militar danés anteriormente desplegado (PAD) con la población general no han incluido datos sobre psicoterapia (en contexto de atención, NdelT) a través de la Defensa o terapia de conversación con el médico general. Este estudio incluyó estas y varias otras fuentes de datos en una comparación exhaustiva de UASM entre el PAD danés y civiles.

Métodos: Se incluyó al personal militar desplegado por primera vez (N = 10.971) que había regresado de una misión a Kosovo, Afganistán, Irak o El Líbano entre enero de 2005 y julio de 2017. Se seleccionó aleatoriamente un grupo de referencia civil emparejado por sexo y año de nacimiento de toda la población danesa no desplegada (N = 253.714). Además, se definió una subcohorte que incluía hombres del PAD y civiles considerados elegibles para el servicio militar. Estas cohortes fueron seguidas en registros médicos militares y registros que cubren los sectores de salud civil primario y secundario de 2005 a 2018, y se compararon las tasas de UASM.

Resultados: Aproximadamente la mitad de las solicitudes iniciales de ayuda del PAD se realizaron a través de la Defensa (49,4%) y el resto a través del sistema sanitario civil. Cuando no se incluyó la búsqueda de ayuda a través de la Defensa, la UASM fue significativamente menor entre el PAD en la cohorte principal durante los primeros dos años (IRR = 0,84, CI del 95%: [0,77, 0,92]) en comparación con los civiles. Cuando se incluyó la búsqueda de ayuda a través de la Defensa, la UASM fue significativamente mayor entre el PAD en comparación con los civiles tanto en los primeros dos años de seguimiento (IRR = 2,01, CI del 95%: [1,89, 2,13]) como posteriormente (IRR = 1,18, CI del 95%: [1,13, 1,23]). En la subcohorte estas diferencias fueron aún más pronunciadas tanto en los dos primeros años de seguimiento como posteriormente.

Conclusiones: La UASM fue mayor entre el PAD danés en comparación con los civiles sólo cuando se incluyeron datos de la Defensa. La inclusión de datos sobre los servicios de salud tanto civiles como militares es necesaria para evaluar el impacto total del despliegue en UASM entre el PAD danés.

KEYWORDS:

• Mental health problems
• military personnel
• military deployment
• veterans
• utilisation of mental healthcare
• mental healthcare services

PALABRAS CLAVE:

• Problemas de salud mental
• personal militar
• despliegue militar
• veteranos
• utilización de la atención de salud mental
• servicios de salud mental

1. Background

The deployment of military personnel to international conflict areas may lead to adverse mental health consequences such as post-traumatic stress disorder (PTSD), depression, and substance abuse (Gadermann et al., 2012; Rhead et al., 2022; Taal et al., 2014; Teeters et al., 2017). However, while much attention has been paid to these individual mental health problems, research comparing the overall mental health of formerly deployed military personnel (FDP) with the general population is scarce and is almost exclusively based on self-reported measures in cross-sectional studies (Forbes et al., 2016; Hoglund & Schwartz, 2014; Rhead et al., 2022; Trautmann et al., 2017). Using self-reports to compare the level of mental health problems across these segments of the population can be problematic due to nonresponse bias for example due to the stigma that for some is associated with reporting mental health problems (Warner et al., 2011). Some researchers have instead taken the approach of estimating the differences in the mental healthcare utilisation (MHU) among FDP and civilians (Aux Analysis AB, 2021; Lyk-Jensen et al., 2016; Madsen et al., 2017; Nissen et al., 2017). MHU is not a direct measure of the level of mental health problems in a population as not all those who experience symptoms of mental health problems seek professional help and as some of those who do receive mental health treatment might not qualify for a psychiatric diagnosis. Despite these limitations, comparing the MHU of FDP and civilians can be useful for revealing important differences in the overall mental health and help-seeking behaviour of both groups.

Denmark has a large range of health registers available for research and offers a unique setting to explore the effects of deployment based on the actual use of healthcare services at the individual level. Furthermore, the Danish Defence has in the last two decades contributed to several NATO and UN missions most notably to Kosovo (6959 deployed) and Lebanon (1066 deployed) which were peacekeeping missions, as well as to Iraq (8304 deployed) and Afghanistan (11,392 deployed) which focused on peacebuilding and training of local forces (Danish Defence Personnel Organisation, 2022). On the latter two missions, many FDP experienced direct combat. Danish FDP are typically deployed for six months. After homecoming, the FDP may – and commonly do – seek help from both civilian and military mental healthcare services both while they are still in service and after they have left the Defence. The Danish Defence offers free psychotherapy for all deployment-related mental health problems. Both FDP and the general population have access to other free mental healthcare services as well, namely treatment at hospital-based mental healthcare clinics and talking therapy from general practitioners (GP). Psychotherapy with a private psychologist is only partly reimbursed by the civilian public healthcare system (CPHS) for a specific set of mental health problems such as depression and anxiety. Prior studies comparing the MHU of formerly deployed Danish military personnel with the general Danish population (Lyk-Jensen, 2022; Lyk-Jensen et al., 2012, 2016; Madsen et al., 2017; Nissen et al., 2017) have, however, neither included register data on ‘talking therapy’ conducted by the GP nor on psychotherapy through the Danish Defence, which appear to be the most commonly used mental healthcare service among Danish FDP (Madsen et al., 2016; Møller et al., 2020). Therefore, the omission of these data could have led to a severe underestimation of the true effect of deployment on the MHU of FDP in the prior comparison studies. Two of the prior studies found lower MHU among FDP compared to the general population (Lyk-Jensen et al., 2012; Madsen et al., 2017); the three other studies found higher MHU among FDP compared to the general population (Lyk-Jensen, 2022; Lyk-Jensen et al., 2016; Nissen et al., 2017). Knowing the actual consequences of deployment on MHU is important to inform soldiers, the Defence as well as decision-makers about the human and societal costs of deployment. Furthermore, obtaining a better overview of where formerly deployed military personnel seek help for mental health problems could be of great value to guide the effort to increase the effect of treatment.

The two main aims of this study were therefore, (1) to examine where Danish FDP seek help for mental health problems following their first deployment, and (2) to compare the incidence of MHU among Danish FDP with a reference group from the general population. We also conducted exploratory analyses to determine whether the frequency of purchase of different types of psychotropic drugs differed between the FDP and the general population.

With respect to aim 1 and based on the existing literature (Lyk-Jensen et al., 2016; Madsen et al., 2016, 2017; Møller et al., 2020), we hypothesised that psychotherapy through the Danish Defence was the most commonly used mental healthcare service among FDP. With respect to aim 2, we hypothesised that when data on the use of psychotherapy through the Danish Defence were included in the outcome, the incidence of MHU would be higher among FDP compared to the general population.

2. Method

2.1. Study population

This study utilised data on FDP deployed in the period 2004–2017. In this period, the Danish Defence only allowed the deployment of personnel who had known ongoing mental health problems in rare circumstances (The Defence Command, 2017). All deployed personnel had undergone pre-deployment mental health screening as part of their regular health assessment. Further formal and informal health screening took place during military training in the months before the first deployment. Therefore, at the time of the first deployment Danish military personnel were generally healthier than a randomly selected subset of the general Danish population that includes a much broader selection of people with anything from no health problems to very severe health issues (Lyk-Jensen et al., 2016; Madsen et al., 2017). This difference in health due to health requirements in the military has been termed the ‘healthy soldier effect’ and has been observed across military and civilian populations in several countries (Kang & Bullman, 1996; Pethrus et al., 2022). These underlying differences between military personnel and the general population led to the creation of the following two cohorts.

2.1.1. The main cohort

The main cohort comprised male and female first-time deployed military personnel who had returned from a mission with the Danish Defence to Afghanistan, Iraq, Kosovo or Lebanon between January 2005 and July 2017, and a reference group randomly drawn by Statistics Denmark from the entire Danish non-deployed population, matched by sex and birth year. During the follow-up period, less than 1% of the individuals in the reference group were employed on a military contract for longer than a year (maximum military service duration for conscripts) and none in the reference group were deployed to an international conflict zone. Therefore, we henceforth term the reference group ‘civilians’. Twenty-five civilians were matched for each FDP. After being matched, each civilian was assigned two dates that corresponded to the deployment date and the homecoming date of the veteran with whom they had been matched. These dates are henceforth termed the ‘deployment date’ and the ‘index date’ for both FDP and civilians. Civilians who did not reside in Denmark on the index date were excluded. Furthermore, to ensure greater similarity between FDP and civilians and to compare the incidence of MHU between the two, both FDP and civilians were excluded if they had used mental healthcare services in the six months prior to the deployment date.

2.1.2. The sub-cohort

To further counter the selection bias of the healthy soldier effect and obtain a more comparable reference group, a sub-cohort was defined that consisted of only those FDP and civilians of the main cohort, who had been deemed fully eligible for military service by the conscription board. As the Danish draft is only compulsory for men, the sub-cohort included no females. Finally, as data on eligibility were not available before 1987 and conscription took place from age 18, individuals born before 1969 were excluded from the sub-cohort.

2.2. Data

Data on exposure, outcome and covariates were retrieved from the Danish Defence, Statistics Denmark or the Danish Health Data Authority. Nearly all data were linked to each person using the unique Danish personal identification number that each individual receives at birth or immigration. However, a part of the data obtained from the Danish Defence was linked using full name and birthdate (see details below). Data on eligibility for military service were retrieved from the Conscription Register of the Danish National Archives. Migration data was obtained from the Register of Historical Migrations. See Supplementary Table S1 for a full overview of the included registers.

2.2.1. Exposure

The main exposure was whether one belonged to the civilian reference group or to the FDP group. The FDP were identified in the Deployment Database of the Danish Veterans Centre.

2.2.2. Outcome

The outcome of this study was the incidence of MHU. MHU was measured as use of either (1) talking therapy with a GP; (2) psychotherapy with a private psychologist that is partly reimbursed by the CPHS; (3) consultations with a private psychiatrist that is reimbursed by the CPHS; (4) purchase of psychotropic drugs, here including sleep-promoting medicine; (5) psychiatric main diagnoses given in psychiatric wards; or (6) psychotherapy paid for by the Danish Defence provided by either a private psychologist or by a psychologist at the Danish Veterans Centre. The outcome of the analyses comparing FDP with civilians (aim 2) used a combined outcome that included all six types of MHU (0 = no MHU, 1 = MHU).

2.2.2.1. Primary and secondary health sector

Data on outcomes 1–3 were obtained from the Danish National Health Service Register (Andersen et al., 2011), which contains data on all primary health sector services entitled to reimbursements by the CPHS. Data on the fourth outcome, the purchase of psychotropic drugs, were obtained from the Danish National Prescription Register (DNPR), which contains data on all drugs that have been purchased with a prescription in Danish pharmacies since 1995. In Denmark, all purchases of psychotropic drugs and sleep-promoting medicine require prescriptions from a medical doctor.¹ The DNPR covered 96.6% of the use of psychotropic drugs in Denmark during the follow-up period, with the rest issued in hospitals (The Danish Health Data Authority, 2022).

Data on the fifth outcome, psychiatric main diagnoses given at psychiatric wards, were retrieved from the Danish National Patient Register, which includes all diagnoses given at Danish psychiatric wards since 1995. The specific diagnostic and reimbursement codes used in the above-mentioned registers are shown in Supplementary Table S2.

2.2.2.2. Psychotherapy through the Danish Defence

Data on the sixth outcome, psychotherapy in or through the Danish Defence, were retrieved from two different sources:

1. For the period January 2005 to November 2010, the medical records on treatment by the Defence’s own psychologists have largely been deleted. For this period, this study therefore only includes information about referral to one of the private psychologists outside the Defence. This information was retrieved from the Danish Veterans Centre’s list of clients starting treatment with a private psychologist paid by the Danish Defence. This was the only data source that did not have information on the Danish person's identification number but were linked using full names and birth dates. The first treatment date was calculated from the sequence of referrals of clients during that year. Details of the calculation can be seen in Supplementary Table S2. The exact client volume that the employed psychologists had in the period is unknown. However, as the employed psychologists only provided psychotherapy in Copenhagen during this period, we expect that at least half of the FDP were referred to the private psychologists in these years.

2. For the period November 2010 to December 2018, data on the use of psychotherapy were retrieved from the electronic client record system ‘Cosmic’, which includes specific treatment dates for clients who have received treatment from a psychologist employed by the Danish Defence. For clients that have been referred by the Danish Defence to treatment at a private psychologist (also paid by the Danish Defence), only reimbursement dates, generally covering reimbursements for the latest month but occasionally up to the latest year, have been recorded.

During the study period, there have been different mandatory counselling/interview sessions with a psychologist shortly after homecoming. These sessions were not included in the outcome. During such sessions, the psychologist could suggest that the FDP seek help. If symptoms were severe and clearly related to the deployment, direct referral to psychotherapy through the Defence could take place. However, the typical procedure has been for FDP to self-refer to the Danish Veterans Centre and only those who in a subsequent intake session are assessed to have deployment-related symptoms of mental health problems are offered treatment (Folke et al., 2021).

Psychotherapy through the Defence is a service only offered to FDP and their next-of-kin. However, as psychotherapy of next-of-kin through the Defence is only granted for issues related to the deployment of an FDP, a decision was made to exclude treatment of next-of-kin from the outcome.

Danish law requires medical record-keeping for both medical doctors and psychologists. Furthermore, there are additional financial incentives for both general practitioners and psychologists to register their activities. Therefore, apart from the missing data from the psychologists that were employed in the Defence from 2005 to 2010, we expect the amount of missing data in the outcome to be very low.

2.2.3. Covariates

The following potentially confounding covariates were included in the analyses: sex, age group, marital status, ethnicity, socioeconomic status (SES), education level, income group, prior MHU, prior transfer income and deployment era. All data on covariates were retrieved through Statistics Denmark, the Danish Health Data Authority and the Danish Veterans Centre. Details on the origin of each covariate can be seen in Supplementary Table S1. Income group was determined by calculating the inflation-corrected mean income over the past five years. Prior MHU was based on the same data sources as the main outcome, with the exception of psychotherapy through the Defence which is only available after the first deployment, and with the addition of data from the Psychiatric Central Register, which contains information on all diagnoses given in inpatient care at Danish psychiatric wards from 1970 to 1994. The included diagnostic codes can be seen in Supplementary Table S2. Prior MHU covered the MHU in the period from birth until six months before the deployment date or the equivalent date for the matched civilians. Data on prior transfer income were obtained from the Danish Register for Evaluation of Marginalization which contains information on all types of public transfer income since 1991. Receipt of transfer income was defined as any type of transfer income such as unemployment benefit, sickness absence from work, or early retirement compensation received in the three months prior to the deployment date. Public financial aid for education and maternity leave did not count as transfer income. The deployment era variable was constructed from the deployment date that was obtained from the Deployment Database of the Danish Veterans Centre. The variable was included to account for any development over time in the awareness or use of mental healthcare services and was divided into approximate 3-5-year periods depending on the number of deployments over the years.

According to the Danish data protection law, studies that only use survey or register data are not required to obtain approval from a research ethics committee. Permission to use medical record data from the Danish Defence was granted by the Danish Patient Safety Authority (case number 3-3013-3162/1).

2.3. Statistical analyses

In order to examine where Danish FDP first seek help (aim 1), we calculated the incidence rates of the first MHU after the index date for both FDP and civilians for all services. Subsequent MHU was assessed only among those who could be followed up for the full three years after they sought help for the first time.

In order to compare the incidence of MHU between the two FDP groups and their respective civilian reference groups (aim 2), we performed multivariable Cox proportional hazards (PH) regression analyses (Hosmer et al., 2008). All individuals were followed up from the index date until the first of one of the following events: MHU, emigration for >1 year, death, or end of follow-up (31 December 2018). The FDP may or may not have parted ways with the Danish Defence during the follow-up. In the main cohort, 2.7% of the FDP and 12.8% of the civilians had missing covariate data and were therefore not included in the multivariable analyses (sub-cohort missing: FDP = 2.2%; civilians = 3.9%). Additional Cox analyses on the main cohort were conducted comparing FDP with civilians among men and women separately. The main analyses in this study followed FDP during subsequent deployments, that is, periods where they could not use the mental healthcare services at home. This strategy was chosen to resemble the actual lives of Danish soldiers where subsequent deployments are relatively common. FDP are not redeployed if they are not considered eligible by both their superiors as well as the physician conducting their regular health assessments. If mental health problems appear during deployment that raise doubts about the operational readiness of soldiers, they are always repatriated. Sensitivity analyses were also conducted where the follow-up of FDP was stopped on the date of redeployment if they were redeployed during follow-up.

The PH assumption was tested by using the cumulative martingale residual test. If the PH assumption was not fulfilled for the main variable of interest, the follow-up time was split into two or more periods to allow for variation in effects over time. If the PH assumption was not fulfilled for the potential confounding covariates, they were controlled for by stratification as category variables. We used a robust sandwich estimator to accommodate for potential dependencies among observations due to matching.

Finally, in order to explore the use of psychotropic drugs among FDP and civilians (aim 3), we calculated the incidence rates of the first use of seven different types of psychotropic drugs for both FDP and civilians.

All Cox regression analyses were conducted using the ‘mets’-package in R version 4.1.0, and incidence rates (IR) were calculated in Stata/MP 17.0.

3. Results

In total, 1.0% (116/11,087) of FDP and 7.0% (19,218/275,561) of the civilians were excluded due to the use of mental healthcare services in the six months prior to the deployment date. More details of the inclusion of individuals in the two cohorts can be seen in Supplementary Figure S1. Overall, 10,971 FDP and 253,714 civilians were included in the main cohort of the study. The sub-cohort consisted of 8490 male FDP and 88,153 male civilians that were all deemed eligible by the conscription board. In both cohorts, FDP and civilians had a maximum follow-up of 14.0 years and a median follow-up of 9.9 years. Redeployment occurred for 39% of the FDP in the main cohort during follow-up and for 40% of the sub-cohort FDP. The IR for deploying a second time was 0.46 per person year.

Table 1 presents the characteristics at or prior to the deployment date of the main and sub-cohorts. In the main cohort, just 6.2% of the civilians and 6.5% of the FDP were women. The most noticeable differences in this cohort were that FDP were more often of Danish origin, had fewer with a very low income (<100,000 DKK) and were less likely to have received transfer income in the three months prior to the deployment date compared to civilians. This is to be expected as many FDP are undergoing training in the Defence in the period before deployment. By the same token, among individuals with low SES, a much larger part of the FDP was employed compared to civilians. In the sub-cohort, the differences between FDP and civilians were somewhat attenuated; however, dissimilarities were still present regarding ethnicity, SES and prior transfer income. In the main cohort, prior MHU was slightly higher among civilians as compared to FDP. In the sub-cohort, the level of prior MHU was similar between the two groups.

Table 1. Characteristics at or before the date of military deployment for formerly deployed Danish military personnel and civilian reference groups in the main cohort and the eligible male sub-cohort.

	Main cohort				Eligible male sub-cohort^a
	Civilian reference group		Deployed military personnel		Civilian reference group		Deployed military personnel
Characteristics	n	%	n	%	n	%	n	%
Sex^b
Male	237.974	93.8	10.257	93.5	88.153	100.0	8.490	100.0
Female	15.740	6.2	714	6.5	0	0.0	0	0.0
Age at start of follow-up^b
<21 years	85.073	33.5	3.622	33.0	36.352	41.2	3.254	38.3
22–26 years	99.187	39.1	4.277	39.0	37.604	42.7	3.732	44.0
27–31 years	26.712	10.5	1.152	10.5	9.154	10.4	985	11.6
32–36 years	10.787	4.3	465	4.2	3.587	4.1	368	4.3
37–41 years	8.086	3.2	359	3.3	1.113	1.3	120	1.4
≥42 years	23.869	9.4	1.096	10.0	343	0.4	31	0.4
Marital status
Unmarried	218.380	86.1	9.325	85.4	83.094	94.3	7.884	93.2
Married	35.283	13.9	1.600	14.6	5.056	5.7	576	6.8
Ethnicity
Danish	208.211	82.3	10.740	98.0	83.716	95.0	8.313	98.0
Western immigrant	17.102	6.8	96	0.9	435	0.5	63	0.7
Non-western immigrant	27.730	11.0	126	1.1	3.984	4.5	109	1.3
SES^c
High	14.050	5.5	663	6.0	3.594	4.1	215	2.5
Medium	18.335	7.2	628	5.7	5.232	5.9	489	5.8
Low and employed	116.359	45.9	8.433	76.9	42.282	48.0	6.686	78.8
Low and on transfer income	104.890	41.4	1.247	11.4	37.045	42.0	1.100	13.0
Education
10th grade of less	78.024	32.7	3.150	28.9	24.155	27.5	2.522	29.9
High school	65.378	27.4	3.639	33.4	32.547	37.1	3.095	36.7
Vocational education	61.995	26.0	2.617	24.0	21.588	24.6	1.974	23.4
Higher education: bachelor/less	23.845	10.0	896	8.2	7.191	8.2	589	7.0
Master degree/PhD	9.453	4.0	590	5.4	2.276	2.6	259	3.1
Mean income last 5 years
≤100,000 DKK	17.213	7.7	246	2.3	5.027	5.9	212	2.5
100,001–200,000 DKK	95.538	42.7	3.591	33.5	35.548	41.9	2.934	35.3
200,001–300,000 DKK	87.133	38.9	5.561	51.9	35.312	41.6	4.274	51.4
300,001–400,000 DKK	18.237	8.1	1.079	10.1	6.919	8.1	755	9.1
≥400,001 DKK	5.866	2.6	228	2.1	2.115	2.5	148	1.8
Prior MHU^d
No	210.815	83.1	9.470	86.3	78.408	88.9	7.518	88.6
Yes	42.899	16.9	1.501	13.7	9.745	11.1	972	11.4
Prior transfer income^e
No	213.586	84.2	10.788	98.3	78.507	89.1	8.372	98.6
Yes	40.128	15.8	183	1.7	9.646	10.9	118	1.4
Deployment era^f
2004–2006	70.541	27.8	3.014	27.5	25.439	28.9	2.395	28.2
2007–2009	90.870	35.8	3.921	35.7	30.996	35.2	2.981	35.1
2010–2012	60.418	23.8	2.644	24.1	20.695	23.5	2.023	23.8
2013–2017	31.885	12.6	1.392	12.7	11.023	12.5	1.091	12.9

Notes. ^aThe eligible male sub-cohort included only males born after 1969 who were deemed fully eligible for military service by the conscription board.

^b The differences observed in sex and age at the start of follow-up are due to the exclusion of individuals who died or emigrated before the index date or who had used mental healthcare services in the six months prior to the deployment date.

^c ‘High SES’ covers the self-employed that have employees, employees with management responsibilities, and employees that have a high level of skills. ‘Medium SES’ covers self-employed without employees, self-employed with no indication of employees, co-working spouse and employees that have a medium level of skills; ‘Low and employed SES’ covers employees that have a basic level of skills and ‘other employees’. ‘Low and on transfer income’ covers transfer recipients. Less than five controls in the sub-cohort had missing values for SES. Values were randomly imputed for these.

^d Covers any mental health service utilisation from birth until six months before the deployment date.

^e Prior transfer income covers any type of transfer income such as unemployment benefit, sickness absence from work, or early retirement compensation (except public financial aid for education and maternity leave) received in the three months prior to the deployment date.

^f For the reference group, the deployment era is the era within which the FDP that they were matched with were deployed.

3.1. Where do Danish FDP seek help?

A total of 49.4% of the initial help-seeking of FDP in the main cohort took place at the Danish Defence. Only 6.2% of the initial help-seeking by FDP took place in the secondary health sector; the remaining 44.3% occurred in the primary health sector. Figure 1 illustrates where the FDP from the main cohort initially sought help after homecoming (left column of boxes) and where they subsequently sought help, if they did from another health sector during the three years after the initial help-seeking (right column of boxes). In the figure, only FDP that could be followed for a full three years after the initial help-seeking are included. Among those registered as initially seeking help from the primary health sector, only 27.0% subsequently sought help from the secondary health sector or from psychologists through the Danish Defence.

Figure 1. Routes of help-seeking among 2848 formerly deployed Danish military personnel from the main cohort who had sought help during follow-up and who could be followed for 3 years.

Notes. ^aA total of 33 of the 1127 persons had a prescription for a psychotropic drug for which it was unclear whether the doctor who wrote the prescription belonged to the primary or secondary health sector (origin of prescription variable ‘RECU’ codes 0990019 or 0990027 or missing). Since the vast majority of prescriptions for psychotropic drugs originate from the primary health sector these 33 prescriptions were also interpreted as originating from the primary health sector.^bA total of nine of the 358 persons had a prescription for a psychotropic drug for which it was unclear whether the doctor who wrote the prescription belonged to the primary or secondary health sector (origin of prescription variable ‘RECU’ codes 0990019 or 0990027 or missing). Since the vast majority of prescriptions for psychotropic drugs originate from the primary health sector these nine prescriptions were therefore also interpreted as originating from the primary health sector.^cFive or less than five of the 73 persons had a prescription of a psychotropic drug for which it was unclear whether the doctor who wrote the prescription belonged to the primary or secondary health sector (origin of prescription variable ‘RECU’ codes 0990019 or 0990027 or missing). Since the vast majority of prescriptions for psychotropic drugs originate from the primary health sector, these five or less than five prescriptions were therefore also interpreted as originating from the primary health sector.

3.2. MHU among FDP and civilians

Table 2 shows the incidence rates of first use of mental healthcare services between FDP and civilians. In both the main and the sub-cohort, the initial use of primary and secondary healthcare services was higher among civilians. However, the overall incidence per 1000 person years of MHU was significantly higher in FDP (main cohort IR = 44.9, 95% CI: [43.5, 46.5]) as compared to civilians (main cohort IR = 37.4, 95% CI: [37.2, 37.7]).

Table 2. Incidence rates per 1000 person years for the first use of a mental health service after homecoming among formerly deployed Danish military personnel and the civilian reference groups in the main cohort and the eligible male sub-cohort.

	Main cohort		Eligible male sub-cohort^a
	Civilian reference group	Deployed military personnel	Civilian reference group	Deployed military personnel
Help-seeking type	IR (95% CI)^b	IR (95% CI)^b	IR (95% CI)^b	IR (95% CI)^b
Total use of primary health sector	32.8 (32.6, 33.1)	19.9 (19.0, 20.9)	27.0 (26.6, 27.4)	18.7 (17.6, 19.8)
GP talking therapy	10.5 (10.3, 10.6)	6.8 (6.2, 7.4)	9.4 (9.2, 9.7)	6.5 (5.9, 7.1)
Private psychologist CPHS^c	5.4 (5.3, 5.5)	3.6 (3.2, 4.1)	5.5 (5.3, 5.6)	3.5 (3.1, 4.0)
Private psychiatrist CPHS^d	1.3 (1.3, 1.4)	0.4 (0.3, 0.6)	1.1 (1.0, 1.2)	0.4 (0.3, 0.6)
Primary sector psychotropic drugs	15.7 (15.5, 15.9)	9.1 (8.4, 9.8)	11.1 (10.8, 11.3)	8.3 (7.6, 9.1)
Total use of secondary health sector	4.6 (4.5, 4.7)	2.8 (2.5, 3.2)	3.4 (3.2, 3.5)	2.7 (2.3, 3.2)
Psychiatric ward mental diagnosis	3.3 (3.2, 3.4)	2.1 (1.8, 2.4)	2.6 (2.4, 2.7)	2.1 (1.7, 2.5)
Secondary sector psychotropic drugs	1.3 (1.3, 1.4)	0.8 (0.6, 1.0)	0.8 (0.8, 0.9)	0.7 (0.5, 0.9)
Psychologist through military	0	22.2 (21.2, 23.2)	0	23.8 (22.5, 25.0)
All help-seeking	37.4 (37.2, 37.7)	44.9 (43.5, 46.5)	30.4 (30.0, 30.8)	45.2 (43.5, 46.9)

Notes. ^aThe eligible male sub-cohort included only males born after 1969 who were deemed fully eligible for military service by the conscription board.

^b IR = Incidence Rate per 1000 person years (95% confidence interval).

^c Includes only use of private psychologists that is partly reimbursed by the Danish civilian public healthcare system.

^d Includes only use of private psychiatrists that is reimbursed by the Danish civilian public healthcare system.

Table 3 shows the results of Cox proportional hazard regression analyses comparing time to MHU among FDP and civilians. When data on psychotherapy through the Danish Defence were excluded, FDP in the main cohort had a 19% lower IR of MHU during the first two years of follow-up compared to civilians (IRR = 0.84, 95% CI: [0.77, 0.92]). After the first two years, no differences were found. In the sub-cohort, there were no significant differences between FDP and the civilians before or after two years of follow-up.

Table 3. Incidence Rate Ratios (IRR) of use of mental healthcare services in formerly deployed Danish military personnel compared to the civilian reference groups in the main cohort and the eligible male sub-cohort.

	Main cohort		Eligible male sub-cohort^a
	IRR^b	95% CI^c	IRR^d	95% CI^c
Military personnel vs. reference group excl. Defence data
– During the first two years follow-up	0.84	[0.77, 0.92]***	0.98	[0.88, 1.09]
– Past the first two years of follow-up	0.95	[0.90, 1.01]	1.05	[1.00, 1.11]
Sex: Female	1.67	[1.63, 1.72]***	NA	NA
Military personnel vs. reference group incl. Defence data
– During the first two years follow-up	2.01	[1.89, 2.13]***	2.55	[2.36, 2.76]***
– Past the first two years of follow-up	1.18	[1.13, 1.23]***	1.35	[1.28, 1.43]***
Sex: Female	1.67	[1.62, 1.71]***	NA	NA

Notes. ^aThe eligible male sub-cohort included only males born after 1969 who were deemed fully eligible for military service by the conscription board.

^b IRR = Incidence Rate Ratios. IRRs were adjusted for age group, marital status, ethnicity, socioeconomic status, education level, income group of the past five years, prior mental healthcare utilisation, prior transfer income and deployment era as well as mutually adjusted for each other. Each civilian was assigned the same ‘deployment date’ and ‘deployment era’ as the veteran with whom they were matched. Prior transfer income covers any type of transfer income such as unemployment benefit, sickness absence from work, or early retirement compensation (except public financial aid for education and maternity leave) received in the three months prior to the deployment date.

^c 95% confidence intervals and p-values. * = p-values <.05, ** = p-values <.01, *** = p-values <.001.

^d IRRs were adjusted for age group, marital status, ethnicity, socioeconomic status, education level, income group of the past five years, prior mental healthcare utilisation, prior transfer income and deployment era.

When data on psychotherapy through the Danish Defence were included, the FDP in the main cohort had a two-fold higher IR of MHU during the first two years of follow-up (IRR = 2.01, 95% CI: [1.89, 2.13]) and a 1.18 times higher IR of MHU after two years of follow-up (IRR = 1.18, 95% CI: [1.13, 1.23]). The difference in incidence rates was even more pronounced in the sub-cohort, where eligible male FDP during the first two years of follow-up had a 2.55-fold higher IR compared to eligible male civilians (IRR = 2.55, 95% CI: [2.36, 2.76]), and a 35% higher IR after two years (IRR = 1.35, 95% CI: [1.28, 1.43]).

The sensitivity analyses (Supplementary Table S3) revealed that when time was stopped at a possible second deployment, the difference between FDP and civilians after two years of follow-up when data on psychotherapy from the Danish Defence was included, was completely removed in the main cohort and substantially reduced in the eligible sub-cohort.

MHU was 67% higher among women compared to men regardless of whether they belonged to the FDP or the reference group and regardless of whether data on psychotherapy through the Danish Defence were included or not (Table 3). Supplementary Table S4 shows the differences in MHU between FDP and the reference group for men and women of the main cohort separately. There were tendencies of male FDP seeking less help from the CPHS compared to women. However, when testing for interactions between deployment status (FDP vs. civilians) and sex, no interaction was found (Wald test on data excluding Defence data p-value = .23; Wald test on data including Defence data p-value = .42).

3.3. Psychotropic drug use among FDP and civilians

Table 4 presents the variations in the use of psychotropic drugs in both cohorts. In the main cohort and in comparison with civilians, FDP had a similar or significantly lower use of psychotropic drugs during follow-up. However, in the eligible male cohort, the use of antidepressants, antipsychotics, benzodiazepine derivatives and sleep-promoting medication was significantly higher among FDP, and only the use of drugs for alcohol/opioid dependence was lower among FDP. The largest difference found was in the use of antipsychotics where FDP had an IR of 5.6 (95% CI: [5.1, 6.1]) per 1000 person years compared with an IR of 3.3 (95% CI: [3.2, 3.4]) per 1000 person years among civilians. This dissimilarity was mainly due to the great divergence in the use of quetiapine, which is frequently used in the treatment of major depression and PTSD.

Table 4. Incidence rates per 1000 person years for different types of psychotropic drugs purchased among formerly deployed Danish military personnel and the civilian reference groups in the main cohort and the eligible male sub-cohort.

	Main cohort			Eligible male sub-cohort^a
	Civilian reference group	Deployed military personnel		Civilian reference group	Deployed military personnel
Category of psychotropic drug	IR (95% CI)^c	IR (95% CI)^c		IR (95% CI)^c	IR (95% CI)^c
ADHD	2.3 (2.3, 2.4)	1.3 (1.1, 1.6)		1.8 (1.7, 1.9)	1.5 (1.3, 1.8)
Alcohol/opioid dependence	2.2 (2.2, 2.3)	1.2 (1.0, 1.4)		1.5 (1.4, 1.6)	1.1 (0.9, 1.3)
Antidepressants	13.9 (13.7, 14.0)	11.8 (11.1, 12.5)		10.3 (10.1, 10.6)	11.6 (10.9, 12.5)
– hereof Mianserin	0.7 (0.6, 0.8)	0.6 (0.5, 0.8)		0.4 (0.4, 0.5)	0.6 (0.4, 0.8)
– hereof Mirtazapine	3.5 (3.4, 3.6)	3.2 (2.9, 3.6)		2.4 (2.3, 2.5)	3.3 (2.9, 3.7)
Antipsychotics	4.9 (4.8, 5.0)	5.1 (4.6, 5.5)		3.3 (3.2, 3.4)	5.6 (5.1, 6.1)
– hereof Olanzapine	0.7 (0.7, 0.7)	1.1 (0.9, 1.4)		0.4 (0.4, 0.5)	1.4 (1.1, 1.6)
– hereof Quetiapine	2.8 (2.8, 2.9)	3.9 (3.5, 4.3)		2.0 (1.9, 2.1)	4.3 (3.9, 4.8)
– hereof Risperidone	0.7 (0.7, 0.8)	0.4 (0.3, 0.5)		0.5 (0.4, 0.5)	0.4 (0.3, 0.6)
Anxiolytics	4.6 (4.6, 4.7)	3.3 (3.0, 3.7)		2.9 (2.8, 3.0)	3.1 (2.8, 3.6)
Benzodiazepine derivatives	0.5 (0.4, 0.5)	0.5 (0.4, 0.6)		0.3 (0.2, 0.3)	0.4 (0.3, 0.6)
Sleep-promoting medication	8.3 (8.2, 8.5)	8.0 (7.5, 8.6)		6.2 (6.0, 6.3)	7.5 (7.0, 8.2)
– hereof Melatonin	2.6 (2.5, 2.6)	2.9 (2.6, 3.2)		2.0 (2.0, 2.1)	2.9 (2.6, 3.4)
– hereof Promethazine	1.4 (1.4, 1.5)	1.4 (1.2, 1.6)		1.0 (1.0, 1.1)	1.3 (1.1, 1.6)
– hereof Z-drugs	5.6 (5.5, 5.7)	5.2 (4.8, 5.7)		4.0 (3.8, 4.1)	4.8 (4.3, 5.3)
Use of any psychotropic drugs	23.7 (23.5, 24.0)	19.2 (18.3, 20.1)		17.4 (17.1, 17.7)	18.5 (17.5, 19.5)

Notes. ^aThe eligible male sub-cohort included only males born after 1969 who were deemed fully eligible for military service by the conscription board.

^b The following ATC codes were included in the following categories. ADHD: N06BA01, N06BA02, N06BA04, N06BA07, N06BA09 and N06BA12. Alcohol/opioid dependence: N05BA02, N07BB01, N07BB03, N07BB04, N07BB05, N07BC01, N07BC02 and N07BC51. Antidepressants: N05AN01, N06AA02, N06AA04, N06AA09, N06AA10, N06AA12, N06AA16, N06AA21, N06AB03, N06AB04, N06AB05, N06AB06, N06AB08, N06AB10, N06AF01, N06AG02, N06AX02, N06AX03, N06AX11, N06AX12, N06AX16, N06AX21, N06AX22 and N06AX26. Antipsychotics: N05AA01, N05AA02, N05AA04, N05AB03, N05AB04, N05AC01, N05AD01, N05AD03, N05AD05, N05AE04, N05AE05, N05AF01, N05AF03, N05AF05, N05AG02, N05AG03, N05AH02, N05AH03, N05AH04, N05AL01, N05AL05, N05AX08, N05AX12 and N05AX13. Anxiolytics: N03AX16, N05BA01, N05BA04, N05BA06, N05BA08, N05BA12, N05BB01 and N05BE01. Benzodiazepine derivatives: N05CD02, N05CD03, N05CD04, N05CD05, N05CD06, N05CD08 and N05CD09. Sleep-promoting medication: N05CF01, N05CF02, N05CF03, N05CH01 and R06AD02. Z-drugs: N05CF01, N05CF02 and N05CF03.

^c IR = Incidence Rate per 1000 person years (95% confidence interval).

4. Discussion

Firstly, this study aimed to examine where Danish FDP seek help for mental health problems following their first deployment. As hypothesised, the most common place to seek help was through the psychologists at the Danish Defence where roughly half of the FDP first sought help. Most FDP did not go on to seek help elsewhere once they had sought help in either the primary or secondary health sector or from the Defence. Secondly, we aimed to compare the incidence of MHU among FDP with that among civilians. As hypothesised, when data on psychotherapy were included, a significantly higher MHU was observed after homecoming from military deployment among FDP compared to the civilian reference group. The differences were even larger in the sub-cohort where eligible male FDP were compared to eligible male civilians. The MHU among FDP was especially elevated during the first two years after homecoming. However, when data on psychotherapy through the Danish Defence were not included in the analyses, the MHU was significantly higher among civilians during the first two years in the main cohort, and no significant differences were observed in the sub-cohort. Lastly, we aimed to explore whether FDP used some types of psychotropic drugs more or less compared to civilians. In the main cohort, FDP had either a lower or similar use of all types of psychotropic drugs. Higher medication use among FDP was only observed in the sub-cohort where FDP had a significantly higher use of psychotropic drugs commonly used for depression and PTSD.

The heterogeneous findings of this study clearly demonstrate how study design choices may influence the results obtained. Both the exclusion of non-eligible individuals in the sub-cohort and the inclusion of data on psychotherapy through the Defence had important effects on the results when comparing FDP with civilians. The exclusion of non-eligible individuals in the sub-cohort, made the civilians more comparable to the eligible FDP at baseline, indicating that the sub-cohort results are based on a more accurate comparison. Furthermore, the substantial amount of MHU taking place through the Defence means that the inclusion of data from the Defence is necessary to understand the full implications of military deployment on post-deployment MHU.

Our finding that the IRR of MHU after two years’ follow-up was reduced in the sensitivity analyses, where follow-up was stopped at redeployment, could both be due to the effect of subsequent deployments not being included in the sensitivity analyses as well as the fact that the sensitivity analyses may not factor in deployment-related mental health problems with a delayed onset among those who redeployed. A meta-analysis by Xue and colleagues found the risk of PTSD to increase with the number of deployments (Xue et al., 2015); however, our findings are the first to indicate a relation between number of deployments and MHU among Danish FDP as two prior Danish studies found no such relation (Lyk-Jensen et al., 2012; Nissen et al., 2017).

The five prior Danish studies comparing the MHU among FDP and civilians produced conflicting results concluding either that FDP had lower or higher MHU compared to civilians (Lyk-Jensen, 2022; Lyk-Jensen et al., 2012, 2016; Madsen et al., 2017; Nissen et al., 2017). However, results from two of the studies offered some possible explanations as to why these prior findings might be so different. The first study (Madsen et al., 2017) found a significantly lower incidence of MHU among deployed to peacekeeping missions in the Balkans compared with deployed to combat missions in Afghanistan and Iraq. The differences in the findings of prior studies may therefore be related to the different levels of combat exposure experienced on different types of missions. Combat exposure has also been found to be an important risk factor for post-deployment mental health problems in other major international studies (Smith et al., 2008; Stevelink et al., 2018). The second study (Lyk-Jensen, 2022) found a general increase in MHU among both FDP and eligible civilians over time, but that the increase in MHU was higher than that observed among eligible civilians. Hence, levels of help-seeking may differ over time among both FDP and civilians and therefore, depending on the period studied, the results may differ.

The only study outside of Denmark that has analysed differences in MHU among FDP and civilians is a Swedish study that found FDP to have comparable and, in many respects, lower MHU than civilians (Aux Analysis AB, 2021). The lower levels of help-seeking among Swedish FDP might be attributed to possible lesser levels of combat exposure on the missions that Sweden has participated in and a more limited offer of specialised veteran mental health care in Sweden (Sundgren, 2019).

The higher MHU observed among FDP when psychotherapy through the Defence was included, naturally raises the question of whether this can be translated to there being higher levels of mental health problems among the FDP. However, this may not necessarily be the case. The high MHU observed among the FDP does point to a great treatment need in the group – especially during the first two years after homecoming. However, psychotherapy through the Danish Defence is a free-of-charge life-long offer for all Danish FDP. The personnel are made aware of the service at multiple times before, during, and after deployment, and personnel with severe symptoms have been referred to treatment directly from the mandatory counselling/interview sessions that have taken place shortly after homecoming. These efforts to ensure the availability of treatment and awareness hereof among FDP have likely led to a higher proportion of help-seekers among FDP with mental health symptoms. Despite the already existing free healthcare services available to civilians, underutilisation of mental healthcare services among civilians is still common (Weye et al., 2023). It is therefore likely that the civilians would have had higher than observed levels of MHU if they also had access to an additional free mental healthcare service.

Although one of the key findings from this study was the importance of exploring help-seeking via the Defence when examining MHU among Danish FDP, approximately half of the help-seeking did take place through civilian services, which are generally accessed through the GP. Furthermore, a large part of the MHU that began in the primary health sector did not lead to the use of specialised mental healthcare services over a period of three years. This could be a cause for concern if some of the deployment-related issues the FDP might present with were not referred to specialised treatment as deployment-related issues such as PTSD generally require specialised treatment. However, we had no information on whether or not the mental health problems the FDP sought help for actually were deployment-related. Therefore, further research is needed that can shed light on the FDP’s use of general practice and the type of help they receive there.

4.1. Strengths and limitations

This study had several important strengths. First, the study included all available register data on both military and civilian mental healthcare services. To our knowledge, this makes it the most comprehensive register-based comparison of MHU between FDP and civilians. Second, by using register data, many potential biases such as recall bias, social desirability bias, or nonresponse bias, were avoided. Finally, by excluding all individuals who were not deemed eligible by the conscription board, we were able to select a more comparable reference group for young male military personnel in the sub-cohort.

This study also had some limitations. First, we had no data on the use of private psychologists who were not reimbursed by the CPHS, or on professionals not approved for reimbursements by the CPHS, such as coaches or psychotherapists with no formal university education. This lack of data could have led to either over- or underestimations of the difference in MHU between FDP and civilians, depending on which group used these professionals the most. Second, we had no data on the use of psychologists employed within the Danish Defence in the period 2005–2010. The lack of these data means that the overall amount of MHU of FDP is likely to be underestimated in this study. Third, we followed the FDP and civilians for a maximum of 14.0 years and 9.3% of the FDP were followed up for less than five years which may not be long enough to capture all deployment-related help-seeking. Further research is needed to elucidate whether there are differences in MHU among Danish FDP and civilians outside this timeframe. Finally, we had no data to adjust for differences in adverse childhood experiences between FDP and civilians that did not result in prior MHU.

In conclusion, our findings demonstrated that when data on psychotherapy through the Defence were not included, IRR of MHU were similar or lower among FDP compared to Danish civilians. When data on psychotherapy through the Defence was included, the Danish FDP initiated significantly more MHU after homecoming compared to Danish civilians, indicating a great treatment need especially during the first two years after homecoming. Future studies should address the shortcomings of this and prior register-based studies, especially regarding FDP’s use of the GP and self-paid psychotherapy.

Acknowledgements

We thank statistician Thomas Scheike for guidance in performing time-dependent Cox regression analyses in R.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Supplementary table S1 shows the origin of the data included in the study. Research projects that fulfil the requirements of the data controllers may retrieve these data.

Additional information

Funding

This research was funded by the Danish Veterans Centre through the 2018–2023 defence agreement.

Notes

1 With only one exception: Before 2014, Promethazine could be purchased over the counter but since then a prescription has been required.